But how can one be certain that, for example, keeping cats might cause multiple sclerosis (perhaps because of some transmissible virus) rather than, for example, that those with multiple sclerosis might be more likely to keep cats as company? The possibilities of the case-control study for investigating the causes of disease, given the precedent of smoking and lung cancer, are immense, but so is the danger of drawing false inferences and conclusions. How can one tell?
Bradford Hill formulated a set of criteria that must be fulfilled. They are illustrated here with the example of smoking and lung cancer.
1The correlation must be biologically plausible: there are cancer-inducing agents in tobacco which, when brought into contact with lung tissue, could cause the disease.
2The correlation must be strong: the death rate from lung cancer in cigarette smokers is twenty-five times higher than in non-smokers.
3The correlation must reflect a biological gradient: the more cigarettes that are smoked, the higher the risk of lung cancer.
4The correlation must be found consistently: thirty-six separate studies examining the relationship between smoking and lung cancer have found a positive correlation.
5The correlation must hold over time: as cigarette consumption has steadily increased it has been paralleled by a rise in incidence of disease.
6The association must preferably be confirmed by experiment. If smoking causes lung cancer, then the experiment of stopping smoking should reduce the risk, and the longer the time since stopping smoking, the lower that risk will be.
Bradford Hill concluded:
Here then are different view points all of which we should study before we cry ‘causation’. None can bring indisputable evidence for or against the cause-and-effect hypothesis [but] what they can do, with greater or lesser strength, is to help us to make up our minds on the fundamental question – is there any other way of explaining the set of facts before us? Is there any other answer equally, or more, likely than cause-and-effect?
There is nothing here other than the application of elementary rules of logic and, as with the concept of the ‘controlled trial’ and the ‘case-control study’, the notion is disarmingly simple: scientific hypotheses about the causes of disease require not single facts, but that those facts be internally coherent, that they ‘hang together’. If these criteria are not fulfilled, then the theory is, by definition, incoherent and must fall. In the following decade Bradford Hill’s cautious and logical views were ignored as epidemiologists claimed to have found that virtually every aspect of people’s lives was implicated in some illness or other, generating both anxiety and confusion in the public mind. But that is another story.
4
1952: CHLORPROMAZINE AND THE
REVOLUTION IN PSYCHIATRY
Serious psychiatric illnesses such as schizophrenia are usually perceived as having ‘something to do’ with the chemistry of the brain, an abnormality of one or other of the many chemicals (or neurotransmitters) that transfer ‘messages’ from one nerve to another, and some of whose names may be vaguely familiar – noradrenaline, acetylcholine and dopamine. Similarly modern drugs – such as chlorpromazine – are presumed to work by correcting these chemical abnormalities.
This schematic view of contemporary psychiatry is plausible enough but incorrect, and importantly so, for it conceals the truly extraordinary nature of the therapeutic revolution in psychiatry in the post-war years. In just over ten years – the decade of the 1950s – six entirely new types of drug were introduced into psychiatric practice and remain its mainstay today. But their discovery was not based on a scientific knowledge of brain chemicals, which was at the time extremely primitive. Rather the drugs came first, being discovered for the most part by chance, preceding by several years the identification of their effect on neurotransmitters.
But that is not all, for even though drugs like chlorpromazine and the antidepressants that followed soon after were subsequently found to alter the chemistry of the brain by boosting or blocking the action of different chemicals, the underlying ‘problem’ – what actually is happening in the brains of the mentally ill – remains, despite an extraordinary amount of research, unknown. Thus it subsequently became clear that chlorpromazine blocks the activity of the neurotransmitter dopamine, from which one might reasonably infer that, to have such a beneficial effect, schizophrenia must in some way be associated with an excess of dopamine in the brain. But as far as modern science can tell – and there are some fairly sophisticated ways of finding out – the dopamine systems in the brains of schizophrenics appear to be completely normal. The current state of medical knowledge about severe mental illness can thus be summarised as follows: we know that a handful of drugs discovered by accident almost fifty years ago are effective in relieving the symptoms of schizophrenia and depression; but why they work, the nature of the abnormal changes in the brain they correct and especially the causes of psychiatric illness remain a mystery.
In 1953, as a newly appointed psychiatrist to Fulbourn Mental Hospital just outside Cambridge, Dr David Clark’s tasks included a weekly visit to the ‘chronic’ ward housing the long-stay (many for a lifetime) patients:
I was taken in by someone who had a key to unlock the door and lock it behind you. The crashing of the keys in the lock was an essential part of asylum life then just as it is today in jail. This led into a big bare room, overcrowded with people, with scrubbed floors, bare wooden tables, benches screwed to the floor, people milling around in shapeless clothing. There was a smell in the air of urine, paraldehyde, floor polish, boiled cabbage and carbolic soap – the asylum smell. Some wards were full of tousled, apathetic people just sitting in a row because for twenty years the nurses had been saying ‘sit down, shut up’. Others were noisy. The disturbed women’s ward was a phantasmagoric place. The women were in ‘strong clothes’, shapeless garments made of reinforced cotton that couldn’t be torn. Many of them were in ‘locked boots’ which couldn’t be taken off and thrown. There was nothing moveable. There were knives which were taken in and counted up at every meal. The women all had their hair chopped off short giving them identical wiry grey mops. As soon as you came in they’d rush up and crowd round you. Hands would go into your pockets grabbing at you, pulling at you, clambering for release, for food, for anything until they were pushed back by the sturdy nurses, who shouted at them to sit down and shut up. At the back of the ward were the padded cells, in which would be one or two naked women, smeared with faeces, shouting obscenities at anybody who came near. Then there were the airing courts. Grey, big courts, paved with tarmac surrounded by a wall 12 foot high and a hundred men milling around. A few of them walking, some running, others standing on one leg, posturing, the urine running out of their trouser legs. A couple of bored young male nurses standing on ‘points duty’, looking at them ready to hit anybody who got out of line, but otherwise not doing anything. A scene of human degradation.1
In the same year, in a comparable scene of human degradation in Birmingham’s Winson Green Hospital, Joel Elkes, Professor of Experimental Psychiatry, was studying the effect of a new drug, chlorpromazine, which had shown promising results in acute schizophrenia but whose effects up until this time had not been assessed in the chronic, ‘burned-out’, no-hope patients in the back wards. ‘Our limited aim’, wrote Professor Elkes, was ‘to determine the usefulness of chlorpromazine in the overactive chronic psychotic patient in the crowded disturbed wards of the mental hospital.’ Among those given the new drug was a 32-year-old schizophrenic man who had been in hospital for six years:
His behaviour was greatly disturbed by terrifying visual and auditory hallucinations. He referred to them as ‘bogies’ and, when present, they occupied his whole attention. He spent a great deal of time writing inconsequential sentences, or drawing to ‘ward off the bogies’, and he would often shout abuse at them, banging the furniture and marching about the ward. His sleep was disturbed and he required sedation nearly every night. Socially he was very
withdrawn and solitary. After three weeks on chlorpromazine he gradually became more accessible and friendly. He worked on the ward and took charge of the Christmas decorations. Eventually he was able to attend occupational therapy for the first time since his admission, where he practised his talents for painting. He received no medication apart from chlorpromazine and had only occasional restless nights despite this. Sporadically he would shout at his hallucinations but he said that ‘the bogies do not worry him so much’. He relapsed when on placebo tablets.2
So here was a man, one of the vast army of the chronically mentally ill in the backwater of the large mental hospitals, who would have been lucky to receive any form of medical attention from one year’s end to the next, but who after only three weeks of taking chlorpromazine was able to ‘take charge of the Christmas decorations’. Elkes could not have appreciated the full significance of such a result but in retrospect it is obvious. If a drug can so effectively transform for the better the mental state of someone with chronic schizophrenia, then perhaps asylums and locked wards are unnecessary; perhaps it might even be possible for their inhabitants to return to live in the community. Chlorpromazine ‘tore through the civilised world like a whirlwind and engulfed the whole treatment of psychiatric disorders’, but to fully appreciate its impact it is first necessary to return to the ‘dark before the dawn’.
Within the spectrum of psychiatric disorders it is customary to distinguish the neuroses, such as anxiety and hypochondriasis, from the psychoses, the severe mental illnesses in which consciousness and perception are impaired, such as schizophrenia and manic depression. It is this latter group who filled the mental hospitals in the 1930s and 1940s in their tens of thousands. Henry Rollin, former medical superintendent of Horton Mental Hospital on the outskirts of London in 1937, recalls:
Some of those with schizophrenia . . . spent their entire day in assumed statuesque postures; others endlessly carried out strange and seemingly meaningless rituals, or rocked rhythmically and tirelessly backwards and forwards in their chairs. Some exhibited ‘waxy flexibility’ where their limbs appeared to be moulded into strange postures which were then maintained for prolonged periods . . .
Comparable but fewer in numbers were those patients deemed to be suffering from the affective psychoses (manic depression) whose major symptoms were to be seen in gross disturbances of mood. In this group there was no shortage of victims of wild, uncontrolled psychomotor excitement who could quite literally die from what was termed ‘exhaustion of mania’. At the other extreme of this were victims of psychomotor retardation in which depression reached such depths as to lead to stupor. This, too, constituted a distinct threat to life as a result of starvation: forced feeding of these unfortunates was a well practised technique.3
Grim the pre-war asylums certainly were, but it is important not to lose sight of the real misfortune of their inmates – the mental suffering caused by the illness for which they had been admitted. Thus the patient with schizophrenia is often fearful, persecuted by frightening hallucinations or the machinations of delusionary others: ‘Fear comes suddenly, chilling and shocking and with it uncertainty and new shadows – shadows with movements and hidden life, the life of the small night-time enemies, rodents, insects, marauders.’4
It is only within this context of quite unimaginable mental suffering, where medicine had nothing to offer other than custodial care and sedation, that it is possible to understand the rise in popularity in the 1930s and 1940s of what now seem crude and often cruel treatments – insulin coma, electric shock (ECT) and psychosurgery. These became known as the ‘physical therapies’, for that is precisely what they were – physical assaults on the patient’s brain in the hope that the trauma would somehow correct its malfunctioning.
The first was ‘prolonged narcosis’, introduced in 1920, where patients were put to sleep for several days with a combination of barbiturate drugs.5 The next, ‘insulin coma’, required patients to be given large doses of insulin which, by lowering the blood sugar, induced a comatose state from which they would be rescued by a large dose of glucose.6 Next they were given a drug, cardizol, that caused them to have epileptic fits,7 and this in turn was replaced by the use of electric shock therapy (ECT) pioneered by an Italian, Ugo Cerletti.8 The last of the physical therapies – lobotomies, where the brain was cut with a knife –was pioneered by a Lisbon neurologist, Egas Moniz.9 The apparent effectiveness of the physical therapies in some patients generated an enormous enthusiasm ‘untainted by the normal requirements of rational scientific scepticism’.10 But they were massively overused, frequently in patients who were quite unsuitable.
They would eventually be eclipsed, not by a psychiatrist or a psychoanalyst or a brain chemist, but by a jobbing French naval surgeon with an inquisitive mind, Henri Laborit. In 1949, while working at the Maritime Hospital in Tunisia, Laborit was investigating ways of treating patients in ‘shock’ who had low blood pressure. Shock may result from a variety of causes: severe blood loss, a failing heart, overwhelming blood infection or major surgery. The shock arising from blood loss can be counteracted with blood transfusion but the cause – and therefore the appropriate treatment – in other situations was at the time unknown. Laborit’s hypothesis, shared by others, was that the trauma of a major operation or overwhelming infection might lead to the release from the cells of chemicals such as histamine (better known as being involved in allergic reactions like hayfever) and that this would produce the fall in blood pressure. If this hypothesis were correct – which it was not – then blocking the release of these chemicals should prevent the development of post-operative shock. Accordingly Laborit gave his patients before and during their operation a cocktail of drugs including the antihistamine promethazine, which blocks the action of histamine (and is similar to the drugs currently used in the treatment of hayfever). Laborit claimed in an article published in 1949 – remarkable for its complete absence of any data – that with this combination ‘we have been able to distinctly influence the development of post-operative problems’.11
More importantly, though, he made an extraordinarily insightful clinical observation about the effects of promethazine. The main drawback of this group of drugs, recognised since their introduction in 1937, was that they caused drowsiness, so it was not surprising that he should note that they had ‘an extremely powerful hypnotic effect’, but he also observed that they had ‘an appreciable analgesic property’ such that he no longer found it necessary to give morphine to deaden the pain following operations: ‘Antihistamines produce a euphoric quietude . . . our patients are calm, with a restful and relaxed face.’ In an interview several years later Henri Laborit elaborated on this observation of euphoric quietude, describing the action of promethazine on the brain as ‘disconnecting’ its functions, resulting in ‘a state of complete calm and tranquillity without depression of mental faculties or clouding of consciousness’.12
In 1950 the drug company Rhône-Poulenc, alerted to the possibility that promethazine might be useful in the treatment of psychiatric disorders, initiated a major research programme. The group of drugs to which promethazine belongs are known as the phenothiazines, and Paul Charpentier, the company’s chief chemist, set out to synthesise as many variations of its molecular structure as possible in the hope of finding one that had the same, or greater, ability to create a sense of ‘euphoric quietude’. The compounds he synthesised were then tested on rats that had learned to climb a rope to avoid an electric shock signalled by the ringing of a bell. One compound in particular, chlorpromazine, left the rats unmoved when the bell was rung.13
Hearing of this, Jean Delay and Pierre Deniker, two leading Parisian psychiatrists, were the first to treat a schizophrenic patient, a 57-year-old labourer, Giovanni A., who had been admitted to hospital for ‘making improvised speeches in cafés, becoming involved in fights with strangers, and walking around the street with a pot of flowers on his head proclaiming his love of liberty’. After nine days on chlorpromazine he
was able to have a normal conversation and after three weeks he was ready to be discharged. This was much better, much quicker, much safer than any response that had been obtained by the physical therapies such as ECT and insulin coma.14 The news then spread to Britain, where Joel Elkes at Birmingham’s Winson Green Hospital – as already described – started to give chlorpromazine to the ‘burned-out’ cases on the long-stay wards on whom at the time no treatment availed.
Chlorpromazine was the first swallow, and in rapid succession over the next few years four other major groups of drugs applicable to the whole spectrum of psychiatric illness – depression, mania and anxiety states – were introduced in exactly the same way, through a combination of chance, shrewd observation and the screening of chemical compounds. Nor indeed could it have been otherwise, for at the time there was simply no perception of how the brain functioned, nor even an inkling of what abnormalities lay behind mental illness and consequently no idea of how these drugs – which appeared to work so well – worked at all.15
Thus in 1955 the Parisian psychiatrists Delay and Deniker, when summarising their experience of treating 1,000 patients, clearly had not the slightest idea of its mode of action. They suggested variously that it might stimulate the sympathetic nervous system, or reduce oxygen metabolism in the brain, or alter the pattern of the brain waves in the same way that occurs during sleep.16 It was not until 1963 – eleven years after Giovanni A. had first been given chlorpromazine – that it was shown to interfere with the action of the neurotransmitter dopamine. It was thus only natural to infer that the underlying problem in patients with schizophrenia was a neurochemical one. Perhaps their brain contained too much dopamine, or dopamine in the wrong place, or the receptors to dopamine in the brain were oversensitive. But this obvious explanation turned out to be incorrect. Neither autopsy studies nor sophisticated scanning techniques have been able to identify or demonstrate any single abnormality of dopamine biochemistry in the brain of schizophrenics (or indeed that of any of the other neurotransmitters).
The Rise and Fall of Modern Medicine Page 8
